House of Lords - Broadband for all - an alternative vision

Broadband for all - an alternative vision - Communications Committee Contents

Chapter 3: The case for open access fibre-optic hubs

46. In this chapter, we will outline our vision for the UK's broadband infrastructure. In doing so, it will be immediately recognisable that our vision does not match the Government's. This is largely because, in considering the current policy, we have found that fundamental questions of first principle were not addressed. We will elaborate below, but it should be noted that despite having, in our view, misjudged their first steps, the Government appear not to have veered too far off course. Accordingly, many of the specific measures we will propose could usefully act as complements to rather than wholesale replacements for those adopted under the current policy, and so throughout this report, we offer recommendations in the hope of contributing to the success of the Government's current efforts. It will be instructive, however, perhaps for future policy makers in this area, to understand how—and why—we part company, in a rather fundamental way, at the outset.

47. Recall, therefore, that the Government's starting point is as follows:

· There are "still some 2 million households who cannot access a good level of broadband,"[42] and additionally;

· It is a matter of "simple fairness … [that] virtually all homes will have access to a minimum level of service."[43]

48. This is a vision of a UK in which virtually all people will have access to an infrastructure with the technical capacity to carry a minimum level of service. Going further, the Government also want as many people as possible to have access to more than the minimum, with the most prominent feature of a higher level of service being the speed at which data can run over the infrastructure. Basic broadband is one thing, but the strategy document makes clear:

"The benefits of superfast broadband have an impact across the whole economy—whether this is through greater scope for tele-working and home-working, which reduces the pressure on the transport network and lowers carbon emissions, or better delivery of public services—such as remote education services."[44]

49. It should also be noted that the Government's strategy conjures a picture of what internet services should look like—or rather, the qualities (particularly speed) which define them. The starting point for their policy, in other words, arises from thinking about broadband primarily as a commercial service proposition, and one in which the general availability of speed is the crucial factor. Then, given its spillover benefits, the question that naturally follows is how to encourage the market, as it stands, to supply it to more people, and as much as possible to an agreeable standard.

50. Certainly, there are important commercial realities to be addressed by policy. In our view, however, another matter should have been considered as well, arguably above all others: what should the UK's broadband infrastructure look like in view of its place as a major strategic asset, at least on a par with the UK's roads, railways and energy networks? This is where policy on broadband infrastructure should start, before equally going on to consider how to get there. A full and rational answer to the question is liable to be rather technical in places. For the purpose of clarity, therefore, we will as much as possible maintain our discussion within the bounds of statements of principle. In doing so, we outline the makings of a coherent vision for the UK's broadband infrastructure, and as such, a more solid foundation from which the Government's policy might have started, and to which, with some judicious tweaking, it might be possible to return.

Our vision for the UK's broadband infrastructure

51. In brief, our vision is of a robust and resilient national network linked primarily by optical connectivity, bringing open access fibre-optic hubs into or within reach of every community. This would allow diverse providers, large and small, to contribute to the reach and resilience of our national connectivity and allow all individuals to benefit from whichever services, including public ones, will run over it in time to come.

52. In addition, in order to realise our vision, our view is that the UK's future broadband infrastructure should deliver the following:

(i) Every community should be within reach of an open access fibre-optic 'hub';

(ii) Every such hub should be fed by ample fibre-optic cable, providing open access to optical links back to the exchange, and back to the public internet—which will not be free, but made available on fair, reasonable and non-discriminatory terms, allowing third parties to build their own local access networks meeting appropriate technical standards, using whichever technologies they choose, from that hub;

(iii) At the very least, we expect a hub to be able to provide backhaul for a wireless network, where there is demand, so that premises would be able to gain access to a wireless internet service from at least one of these hubs—assuming they can afford to do so.[45]

53. We recommend that the Government consider our vision for the UK's broadband infrastructure as set out in this report. As a first step, we recommend that the Government undertake to produce detailed costings of our proposal, not least because our proposal removes the final mile—the most expensive per capita component of the network— from the costs requiring public subsidy.

BOX 1

The open access fibre-optic hub and dark fibre

The open access fibre-optic hub is, in fact, quite a simple idea. It refers to a physical object—in all likelihood a box—situated in the vicinity of a community. Its job is to act as a waystation between that community and the broadband infrastructure that spreads out across the rest of the country. Running into the hub from the wider network would be an ample number of fibre-optic cables, which in the first instance, would be 'dark,' in the sense that no data traffic will yet be running over them. The reason for this may be, for example, that it has not yet been connected in any way to the properties in the community around it. An important feature of the hub, however, is that the dark fibre running into it should be open access; so that anybody is permitted to build a link between a premises in the community and a fibre in the hub by installing their own passive or active electronic equipment in between, and then rent the existing fibre they are connecting to, which extends the connection from the premises out from the hub and onto the wider network. This would enable any type of compatible access network to be built by any local community, SME or infrastructure provider.

54. The idea of the open access fibre-optic hub within reach of every community is not new. In fact, the Government themselves initially had plans analogous to those above, announcing in December 2010 that: "an ambition to deliver a 'digital hub' in every community in the country is at the heart of the... strategy."[46]

55. It should be noted, however, that since then, the idea appears to have drifted somewhat from their plans. In oral evidence, Robert Sullivan, CEO, BDUK, told us:

"I would not say the digital hub concept exists in pure terms, but the philosophy of getting fibre as deeply into the network as we can is still inside the system."[47]

56. In particular, the basis on which the hub can still be seen as a component of the Government's plans is, according to Mr Sullivan, the availability of an on-demand product from BT. He told us: "FTTH on demand from BT is a good example of what we had in mind."[48] For reasons which are made clear in BOX 1, however, the exclusive ability of one provider to build a final fibre link is actually a categorical departure from the idea of an open access fibre-optic hub in which anyone is permitted to build a link between premises in the community and a fibre in the hub. In fact, it well and truly puts the kibosh on the idea. While the Government clearly considered the proposal in general terms early on in their deliberations, it is fair to say now, that it has disappeared from their plans in implementation.

57. We note indications, however, that despite its departure from the original plan, measures which would bring about an end-result with benefits similar to those of the open access fibre-optic hub, are being considered by the European Commission as conditions of the umbrella state aid permission BDUK are seeking for projects operating within its Framework. Interestingly, reports suggest that the European Commission is particularly concerned about the issue of open access and this is causing delay.[49] This is unsurprising insofar as in June 2012, the European Commission launched a consultation, which closes on 1 September, on the revision of the guidelines on public funding for broadband services.[50] This document is instructive, not least because it seems emphatic on the issue of open access.

"... Third parties' effective wholesale access to a subsidised broadband infrastructure is an indispensable component of any State measure supporting broadband.In particular, wholesale access enables third party operators to compete with the selected bidder (when the latter is also present at the retail level), thereby strengthening choice and competition in the areas concerned by the measure while at the same time avoiding the creation of regional service monopolies...

"... Due to the economics of NGAs, it is of utmost importance to ensure effective wholesale access for third party operators. The subsidised network must offer access under fair and non-discriminatory conditions to all operators who request it and will provide them with the possibility of effective and full unbundling. Third party operators must therefore have access to passive and not only active infrastructure. Apart from bitstream access and unbundled access to the local loop, the access obligation should therefore also include the right to use ducts and poles, dark fibre or street cabinets."[51]

58. This would seem to suggest, inter alia, that if the Government were to implement our approach—which, by design, incorporates open access, there would be fewer, if any, state aid complications. In a further strand of European Commission policy, consistent with its state aid consultation, the Commission has recently published its conclusions following consultation on the broadband investment environment. The first conclusion drawn states that:

"Competition needs a level playing field. In particular, alternative players should not have to compete with one hand tied behind their backs: incumbents should not be able to discriminate between their own retail arms and others'. Although often undervalued in today's regulatory practice, securing truly equivalent access by alternative operators to incumbent networks is probably the most important guarantee of sustainable competition, on existing and new networks."[52]

59. We endorse the European Commission's suggestion that open access to dark fibre at the cabinet-level should be introduced as a condition of BDUK's umbrella state aid permission.

60. Accordingly, not least, in order to expedite its own programme, we recommend that the Government incorporate open access to dark fibre voluntarily as a feature of its Framework agreement with suppliers.

61. More broadly, we endorse and invite the Government's view on the European Commission's conclusion on the broadband investment environment that: "securing truly equivalent access by alternative operators to incumbent networks is probably the most important guarantee of sustainable competition, on existing and new networks."

62. The state aid negotiations are a useful illustration of the point made earlier that, while the Government may have made an initial misstep in failing to stop and think strategically about what the UK's broadband infrastructure should look like, perhaps serendipitously, they have not veered too far from the path down which proper consideration of this question would have led them. To that extent, we will be able to make a number of recommendations throughout this report which could relatively easily help put the Government's strategy back on course. For the purposes of clarifying the differences between our vision and the Government's, it will remain more straightforward to think of the open access fibre-optic hub idea as an entirely alternative proposal to the approach being pursued by the Government. Accordingly, we will continue to refer to it as such.

63. With just this brief outline in place, the significance of the differences between our vision and the Government's may be somewhat opaque. In the remainder of this chapter, we will try to make this significance clearer. In straightforward terms, there are three principles behind our vision and the open access fibre-optic hub approach which could usefully be incorporated into the Government's thinking:

· Broadband policy should be driven, above all, by the need to arrest and ultimately eliminate the digital divide, creating the opportunity to unleash its social benefits for all UK citizens;

· It should also be driven by an avowedly long-term, but also flexible view of the infrastructure's future, avoiding prejudicing any particular technology, or falling into the trap of specific speed targets.

· Broadband policy should also strive to reinforce the robustness and resilience of the network as a whole.

We will take these in turn.

Principle 1: Reducing the digital divide

64. The first principle that lies behind our vision is as follows: policy in this area should be driven, above all, by the need to arrest and ultimately eliminate the digital divide, creating the opportunity to unleash its social benefits for all UK citizens. In fact, the Government really have to be aware of two 'divides.' The first separates those communities who can access the infrastructure of their choice from those who cannot. The second separates those who, once they have access to the internet, take it up and use it effectively, from those who do not or cannot.

65. The digital divide with most obvious relevance to the present inquiry is the first one, between 'haves' and 'have-nots,' separating communities who can gain access to the infrastructure they would choose from those who cannot. Of course, the potential benefits of reducing this divide are inestimable, with effects on, among other things, the ability of individuals to work from home, on the ability of socially isolated people to stay in contact, and ultimately the ability of national and local government to provide public services, even to far-flung, remote communities. The Government's own strategy document makes a similar point that: "for example tele-working can:

· Help reduce the barriers to entering the labour force for those groups which may be less mobile (e.g. disabled and parents with child-care responsibilities who wish to work part-time);

· Potentially contribute to the reduction in traffic congestion and carbon emissions;

· Improve work-life balance".[53]

66. In our view, social benefits of this type are the most significant reason for public policy interest in broadband access, and the use of public money is therefore justified to ensure everyone has the opportunity to benefit from them. We were interested to hear from Suvi Lindén, former Finnish Communications Minister and Special Envoy for the Broadband Commission for Digital Development, about the social benefits, particularly in healthcare, that could be realised from enhanced broadband connectivity:

"We just cannot afford to provide the same level of care as we are doing now and that people are used to having in Finland. I have quite often said that, for some municipalities that have these elderly people living, for example, 80 kilometres from the centre of the municipality, 'It is cheaper for you to build up the fibre to this old lady than to take her to live in the city centre in the nursing home.' It costs what it would cost for the municipality to have her living in the public-provided nursing home for two months."[54]

67. It is not clear, however, whether the Government's strategy will eliminate the divide between those communities who can and those who cannot enjoy these benefits. This is because, on the one hand, there is no guarantee that the Government will meet their targets; and on the other, the targets themselves are inherently divisive. In concrete terms, they set a course for a UK in which "virtually all homes will have access to a minimum level of service"[55] and in which "superfast broadband should be available to 90% of people in each local authority area."[56]

68. The divide inherent in these targets is, as noted above, a natural consequence of developing policy in response to the wrong question. With the proposition in mind that policy needs above all to stimulate the market to widen the provision of broadband as a service, the investment challenge is the obvious obstacle. In the stated absence of sufficient Government resources to subsidise provision in all areas which present such a challenge, it then becomes inevitable that some areas will be left behind. This is all the more disappointing for the fact that the communities who would enjoy the greatest uplift in their quality of life from enhanced broadband provision are precisely those likely, on such a policy, to find that it still lies beyond their reach, maintaining if not widening the digital divide.

69. By contrast, if policy development begins by asking what the UK's broadband infrastructure should look like, and with the principle that whatever the answer, it must serve to reduce if not eliminate the digital divide, it will set itself a challenging task, but one which puts the Government on the right track. This is precisely the principle which lies behind our vision for the UK's broadband infrastructure: a national network which brings open access fibre-optic hubs within reach of every community. Over time, this would ensure that all communities are able to benefit from the advantages of access to broadband.

70. It may be helpful, briefly, to explain how the open access fibre-optic hub idea could make this work, and hence why it is an improvement. As things stand, the exact capacity provided to individual premises with relatively poor internet connections is generally a factor of the length and quality of the copper wire that runs between their property and the local exchange. For example, to take speed as an indicative measure, with the latest DSL technology:

· a 700 metre line could achieve 40 Mbps;

· for 2km of copper, the download speed may reach 'up to 5Mbps';

· with more than 3km of copper, the download speed will probably be less than 2Mbps.[57]

71. This degradation in capacity of a copper line means that, at a certain point, capacity is so compromised, that a community would be better served by radio connections or by laying more local fibre. The obvious question that arises is how a new radio mast could be connected back to the existing network, or how fibre connections could be extended out to reach a community. At present, a number of regulatory and proprietary obstacles would stand in the way—we will discuss these and their specific remedies in Chapter 4. At the moment, no one wanting to undertake either of these projects, except the incumbent provider (overwhelmingly BT or Virgin Media), would realistically be able to do so.

72. By contrast, as noted in Box 1, a key feature of open access fibre-optic hubs would be the fact they are precisely that—open access, meaning any third party, not merely the incumbent provider, would be able to use them. This would enable any type of access network to be built by any local community, SME or infrastructure provider to make the final connection between the hubs and people's properties, so long as they meet certain mandated technical standards.

73. To that end, once fibre is driven out to the hub, we anticipate the emergence of a new industry of infrastructure providers who will be able to respond to local demand and build out local access networks accordingly (see Chapter 5). As these emerge, the choice between fibre and wireless technologies in the local access network will naturally be driven by local demand and market forces. Certainly, fibre will represent a premium choice, but where resources cannot stretch to it, wireless links will provide a more affordable option. A wireless network serving over 200 subscribers on Knoydart and the Small Isles, in one of the most remote areas in the UK, is already providing symmetric speeds of up to 30 Mbps. These speeds are currently constrained by the limited capacity of the backhaul connections available, not the local wireless links. The fibre-optic hub idea, therefore, represents a way to reduce significantly if not eliminate altogether the digital divide.

74. It should be noted, however, that we do not see the introduction of the open access fibre-optic hub as a silver bullet. A number of measures, both regulatory and policy-related, would need to be in place to enable its success. We will elaborate on these points later, noting that many of them could, as we have indicated, help to bring the Government's approach closer to ours; it is certainly not too late. For now, what matters is to recognise that policy ideas are available which can make a substantial difference to the digital divide. To bring them into view in good time, however, policy development must start by asking what the UK wants its infrastructure to look like, and what's more, with the principle firmly in mind that any measures undertaken should strive, above all, to reduce and ultimately eliminate the digital divide.

75. Broadband policy should begin from the question: what should the UK communications infrastructure look like?

76. In addition, it should be a fundamental principle of broadband policy that whatever measures are undertaken to enhance or extend its availability, they strive to bring about equality of opportunity to access broadband across all communities in the UK.

77. In this sense, Government policy on broadband should be driven, above all, by the social benefits it can unleash, and the need to arrest and ultimately reduce a damaging digital divide.

Principle 2: Taking a long-term, but flexible view, and avoiding the 'speed trap'

78. The second principle that lies behind our vision, as set out in paragraph 63, is as follows: policy should be driven by an avowedly long-term, but also flexible view of the infrastructure's future, avoiding prejudicing any particular technology, or falling into the trap of specific speed targets. We also note that different factors—whether speed, symmetry, contention, latency, resilience or a combination—will be critical for different users, depending on their needs (see paragraph 15).

79. Consider the Government's goals. By the end of this Parliament in 2015, they are aiming for:

· "providing everyone in the UK with access to broadband speeds of at least 2Mbps";

· "and superfast broadband should be available to 90% of people in each local authority area."[58]

80. First, it has to be said that in a field quite so riddled with jargon and industry branding as broadband, it has been singularly unhelpful for the Government to use a term as vague as 'superfast,' and in so doing, to give it quite such a flexible meaning. By our count, they have used three varying definitions:

· In its early guidance to local authorities, the meaning was clear: "BDUK has defined superfast broadband as having a potential headline access speed of at least 20Mbps";[59]

· Later, in the glossary of its delivery model, "BDUK defines superfast broadband infrastructure as infrastructure capable of delivering speeds higher than 24Mbps, in line with the Ofcom definition";[60]

· More recently, in its guidance on state aid, this definition has shifted to "infrastructure capable of delivering superfast broadband speeds (meaning speeds of 30 Mbps or in any event more than 24 Mbps)";[61]

· Furthermore, the BT Infinity product used to deliver 'superfast' in Cornwall only guarantees a minimum of 15Mbps.

81. Due to this variability, the Government have been forced to maintain 'superfast' as a somewhat moving target:

"All new projects must target delivery of superfast broadband speeds of 30 Mbps or more, which is in line with the EU's superfast/NGA broadband targets. However, due to earlier UK definitions of superfast referring to speeds of more than 24 Mbps, projects already underway will be satisfying the superfast broadband speed requirement if they seek to deliver speeds of more than 24 Mbps."[62]

82. Second, the shifting nature of this target also points to a conceptual problem with the notion of superfast—and any speed target—more generally: the relatively arbitrary differentiation it creates between a particular marker, 'superfast,' on the one hand, and 'basic,' as a minimum level of service, on the other.

83. Speed is a continuum. The Government may have upgraded their definition of superfast to 30 Mbps, but the industry often defines this at higher levels, and sometimes at 100 Mbps or more. Vint Cerf, Google's Chief Internet Evangelist and renowned 'father of the internet,' is often quoted as half-joking that, in the end, "'broadband' is whatever network speed you don't have, yet!"[63]

BOX 2

Definitions of Superfast

84. By planting specific speed targets on this continuum, the Government have been forced to orient their strategy around ways of ensuring that people necessarily get these speeds. This may sound like a virtue, but this early focus on superfast speeds, has, further down the line, in our view, had a serious and compromising effect on the detail of Government policy, regulation and the industrial decisions which have followed from it.

85. The root of these problems is the following: guaranteeing end-users specific speeds as part of policy obliges the UK to build an infrastructure which goes all the way to people's homes and businesses; it forces the Government to focus their investment—and policies to stimulate investment—on the local access network.

86. As reported in Chapter 2, local access or the 'final mile' is the part of the network whose construction presents the most challenging investment case. As a result, infrastructure providers face weak incentives to invest in upgrades which might be in the long-term and wider interest of the UK, but for which there may be no commercial case in the shorter time horizon palatable to their shareholders. It might be worth explaining why this is the case.

87. Recall that the starting point for the Government's policy arises from thinking about broadband primarily as a commercial service proposition, and that then, given its spillover benefits, the question that follows is how to encourage the market to supply it to more people, and as much as possible to an agreeable standard. Given the foregoing discussion, the question the Government—within the scope of this policy thinking—are therefore forced to answer is: how to encourage the market to act in the wider interests of the UK? Of course, they cannot simply make choices for investors. The option which remains is to stimulate the market to act in a responsible way through other means—by providing subsidy for the areas of the country which the market would otherwise ignore. Such a policy, however, assumes that in accepting this subsidy and proceeding with investment, the market will do so in a way that reflects the wider interests of the country—in other words, as we argue below, in a way that makes the infrastructure suitably inclusive, flexible and straightforward to upgrade in the future as needs and demands change and evolve.

88. The market, in current circumstances, is unlikely to act in this way. To extend out the 'final third,' access to backhaul is required. Where the market has failed to generate competition, connections to backhaul are predominantly owned by the incumbent, BT. BT benefits from economies of scale and scope; it can aggregate demand from a range of the segmented markets that fall under the broad umbrella of providing connectivity.[64] These advantages provide BT, inter alia, with the opportunity to take greater risks than its competitors when it comes to its investments, which as we will clarify in Chapter 4, can have the effect of discouraging others from investing at all. In turn, this 'competitive flight' removes the threat of competition and hence the presence of any real pressure or urgency on BT to invest itself—a set of circumstances which may have one of two outcomes: no investment, and no further infrastructure construction at all (this is the market failure associated with the final third, and the problem the Government's policy seeks to address); or investments are made, perhaps stimulated by subsidy, but which may be short term, with no easy upgrade path to accommodate future demand for enhanced capacity, and which will not serve the interests of the UK more widely, as described above.

89. In summary, given the—at present—relatively non-competitive nature of infrastructure provision, if you start from the proposition that policy needs above all to stimulate the market into providing its services more widely, you risk allowing the market to make decisions about a vital component of national infrastructure which are not in the longer-term and wider interests of the UK. Embarking on policy making in such a way moves the focus away from the right starting point: what should the UK's broadband infrastructure look like in the first place, given its status as an asset of strategic, national value?

90. As well as leading to an over-reliance on the market, the Government's failure to start from the right question, and instead focus straightforwardly on the delivery of faster speeds, has created other problems. These are probably best understood in the light of a wider debate which the Government's focus on superfast has generated. Its central question is: what speed of broadband does the UK really need?

BOX 3

The speed debate

The debate ranges widely and often features strong views about the uses to which superfast speeds could be put—and in turn logically sophisticated discourse about whether these uses and services are of the kind which merit Government support for the infrastructure over which they would run. On one side, speed evangelists cite applications such as tele-health, cloud computing, and HD IPTV (Internet Protocol Television) as providing justification for public support. These, it is argued, carry significant positive externalities or 'spillover' effects, and may have considerable impact on the wider UK economy and society. These spillover effects range from the creation of new, 'distributed' industry models, and a whole variety of effects on home-working, socially isolated people and the provision of public services to remote communities. On the other side, sceptics argue that the same behaviours and industries can thrive on an infrastructure that has undergone an evolutionary upgrade, rather than a revolutionary overhaul, and so can be brought about at far lower public cost. In short, there is not yet a 'killer application' (killer app) which puts the case for wider or universal access to superfast speeds beyond all question.

91. Our view is that the debates about speed are somewhat of a red herring. There is justification for the approach of those sceptical about the need for high speeds:

· asking first whether the purposes to which broadband infrastructure will be put create a genuine need for certain speeds;

· and in turn, whether those end uses are sufficiently in the public interest that public subsidy is justified to hasten the construction of the infrastructure they will rely on.

92. Indeed, we endorse this second point and believe strongly that broadband access is socially useful and merits public subsidy. However, definitive answers to these questions are pivotal, only if policy has been oriented around delivering specific speeds to individual end users in the first place. Doing so is flawed for the reasons we have given, as well as because it inevitably forces Government to make dogmatic statements about things which are better left flexible. By contrast, where policy making begins by asking what the UK's communications infrastructure should look like, and with the principle that it should be driven by an avowedly long-term, but also flexible view of the infrastructure's future, it will set itself a challenging task, but one which, in our view, puts the Government on the right track.

93. This is precisely the principle which lies behind our vision for the UK's broadband infrastructure: a national network which allows diverse providers, large and small, to contribute to the reach and resilience of our national connectivity and individuals to benefit from whichever services, including public ones, will run over it in time to come. Again, it may be helpful to explain how the open access fibre-optic hub idea could make this possible.

94. As outlined in Box 1, the significance of the open access fibre-optic hub is that it brings accessible, and flexible points of connection to the wider network within the reach of individuals and communities. Accompanied by changes to the regulatory backdrop of the market discussed in Chapter 4, this would create a fertile environment for competition in the local access network and allow these people to choose (and pay for) the kind of final link or drop which they want themselves—with all of the consequences this may have for upload and download speeds, contention, latency, jitter and so on. This flexibility to build out a local access network sufficient for demand at any given point in time not only eliminates the need for relatively arbitrary mandates for speed in the present, but will also make it possible to meet whatever foreseeable demand there may be in the future.

95. One undeniable feature of the killer app debate is the impossibility of predicting which services will be in demand, and what the needs of society will be in the longer term. Even sceptics concede, however, that we have at our disposal a specification for broadband infrastructure whose technical capacity is so immense, that with it, even the most stretching estimates of our likely needs could be met well into the future. It is called point-to-point FTTP.[65]

96. What is more, the most stretching and long-term estimates are ones we should take seriously and, at a minimum, ensure our infrastructure can flexibly be upgraded to cope with. There are a number of reasons why.

97. First broadband infrastructure is an asset of strategic, national value. It does not itself bring about, but it does enable, meaningful social and economic innovation and change; if we want to be ambitious for ourselves, we have to be ambitious for it, and under any circumstances it should not hold us back. In this respect, we agree with the FTTH Council Europe: "While the future needs of society cannot really be fully anticipated, a network which has theoretically almost infinite capacity is preferable to networks which are already constrained."[66]

98. Second, there is also a basis on which we believe we can, with some confidence, come to a view of the speeds the infrastructure will have to support in the future. Neilsen's law,[67] like Moore's law,[68] extrapolates from past experience to predict the future. It has proved a remarkably accurate predictor of bandwidth capacity and demand. On this basis, we should expect that those who are satisfied with 2Mb/s today will demand ~8Mb/s in 2015 and ~64Mb/s by 2020, and so on. Our infrastructure must be designed to cope with this advancing demand.

99. Finally, the history of the internet suggests that it has enabled new developments, not because it was designed with any of them in mind, but because it offered a very simple, technical capacity for the transfer of information from one place to another which people could then use their creativity and ingenuity to exploit. We would expect greater capacity to unleash a plethora of new ideas and services. As Francesco Caio, author of the 2008 Department for Business, Enterprise and Regulatory Reform report on broadband, put to us in oral evidence: "once you have optical continuity between point A and point B, the internet will do the rest for you."[69]

100. We acknowledge that the local access network presents a challenging investment case and moreover, that point-to-point FTTP, as a gold standard, represents the technical specification with the highest capital outlay required to construct it. There are two options to be considered:

· Given limited resources, policy should be directed at building out an end-to-end infrastructure which is affordable, however likely it may be that in the first instance, some will be left with nothing, and in time, this will itself become redundant and require replacement;

· Given limited resources, policy should be directed at building out an infrastructure to a high specification, not necessarily all the way to people's doors but within their reach, and providing them with the technical and affordable means to connect to it—in whatever way they choose—themselves.

101. The former option appears to have formed the basis of the current policy, while the latter is the one we endorse.

102. Consider in the light of these points, what is happening. The Government's procurement process is widely expected to award Openreach a clear majority of the contracts to upgrade local access networks around the country.[70] [71] In turn, Openreach is planning to use this public funding to invest in types of network and specific technologies which may meet the speed targets set by the Government, but which, looking beyond 2015 are both relatively constrained and liable to necessitate an expensive phase of upgrade, if not outright replacement in the future. We will explain our specific concerns in more detail in Chapter 4, but we identify for now two particular problems which arise from the industrial choices being made in the light of current policy:

· New and unnecessary barriers to the future upgrade of the UK's broadband infrastructure will be erected;

· Availability of enhanced capacity broadband, and any services which this might enable, will not be equal and the digital divide is likely to widen.

103. These preferred choices are entirely understandable as representing the rational choice of an investor under present market conditions with all the relevant pressures that arise from being a public limited company with an existing infrastructure supporting a profitable business. As noted, they are also the natural consequences of a policy developed in response to misconceived considerations—solely about how to stimulate the market into doing more of what it already does.

104. In contrast, properly thinking through questions of first principle brings into view policy routes which can avert these problems. Before considering this, we note that, even within the bounds of the Government's market-stimulation approach, one route to overcoming this challenge might be simply to accept as unrealistic the emergence of any competitive pressure in the market liable to increase the ambition of commercial players' plans. Such an acceptance could, of course, leave the Government in thrall to the commercial interests of such businesses rather than providing them with an ability to direct broadband infrastructure in the wider interests of the UK. For these reasons, this line of analysis, could prompt calls for nationalisation, and these might well have been deafening in a different era. Curiously, and as a matter of fact, we note that there have been calls, not for nationalisation, but for dispensing with competition and handing to Openreach the necessary public money and the entire job of rolling out fibre.[72]

105. We note, however, that one of the advantages of focusing policy around the promotion of open access fibre-optic hubs, as we recommend, is the credible introduction of competitive pressure to invest in local access networks for the long term. Public subsidy should therefore be used to roll out open access fibre-optic hubs to within reach of every community; the local access network, given a reformed regulatory and policy backdrop which we consider in Chapter 4, then becomes a different economic phenomenon to the rest of the network and one in which competition, and indeed community involvement would be newly stimulated. Under such an approach, it would be possible for technical enhancements to be introduced without specific, centralised mandates for speeds or any other quality, for which demand may only exist in the future. With the middle mile within the reach of individuals and communities, such decisions can be made locally. They can, through the operation of the market, choose the kind of final link or drop which they want themselves and upgrade this flexibly as and when they choose to, or there is need to do so.

106. What matters for the network, and therefore for policy is not speed per se, but a simple, long-term assurance that, as innovative new applications emerge—some of which may be core public services—everyone in the UK will have the ability and opportunity to access an infrastructure which means they can benefit from them. To achieve this, a suitably far-sighted vision for a national, strategic asset like the national broadband network is of an infrastructure that provides robust and efficient communications reaching every community in the UK, and which also allows anyone to procure and exploit connectivity—where possible, optical fibre connectivity—between any two or more points in the UK.

107. The focus of the current policy on ensuring specific speeds in end-to-end networks is, however, likely to support investment in technologies which will certainly help carry faster speeds to a proportion of UK citizens by 2015, but which will not do so universally and may confound and positively work against progress to be made beyond 2015 towards an infrastructure capable of supporting whatever technical capacity the UK as a whole might come to need.

108. For these reasons, we believe policy in this area should be driven by an avowedly long-term, but also flexible view of the infrastructure's future, and inasmuch, should avoid falling into the trap of orienting itself around speed targets, which in the main, serve to provide useful material for sloganeers but also obstruct a more sophisticated approach to building the communications infrastructure the UK requires.

109. As this report was being finalised, HM Treasury unveiled a new UK Guarantees scheme with the aim of accelerating major infrastructure investment and providing support to UK exporters.[73] Under this scheme, the Government's intention is to ensure that where major infrastructure projects are struggling to access private finance because of adverse credit conditions, these projects can proceed. Applications can be made to Infrastructure UK, the HM Treasury body focused on prioritising and enabling investment in UK infrastructure. It is thought that around £40 billion of projects could qualify for the provision of guarantees and these projects could come from a range of sectors including transport, utilities, energy and communications. This is, in our view, an interesting development and we think that broadband infrastructure projects would be appropriate beneficiaries of guarantees.

110. We recommend that future broadband policy should not be built around precise speed targets end-users can expect to receive in the short-term, however attractive these may be for sloganeers.

111. In addition, broadband infrastructure policy should be driven by an avowedly long-term, but also flexible view of the infrastructure's future.

112. As an overriding principle, we recommend that Government strategy and investment in broadband infrastructure should always be based on a minimum ten year horizon and possibly beyond.

113. While we acknowledge the presently elusive nature of a 'killer app,' we believe there is a clear need for the Government to state in explicit terms a long-term vision for a pervasive, robust and resilient broadband infrastructure, central to national policy and infrastructure planning.

114. We anticipate and recommend that policy should be ultimately directed towards universal, point-to-point FTTP as this is a technology not only able to accommodate current demand, but at current rates of growth, will be able to accommodate the UK's bandwidth demands for many decades to come.

115. In this sense, we recommend that the Government should set out an even bolder vision for broadband policy than is currently the case.

116. Given the impossibility, with current constraints on resources, of rolling out universal point-to-point FTTP, we recommend that Government policy should, as an intermediate step, aim to bring national fibre-optical connectivity—which would include, as a minimum, fully open access fibre backhaul—within the reach of every community. This will provide the platform from which basic levels of service can be provided to all, and an improved service where there is sufficient demand.

117. As a point of principle, we believe it is incumbent on the Government to ensure that policy and regulation in the interim guarantee that there is a clear path from any intermediate steps which may be taken to the roll-out of point-to-point FTTP and that, crucially, these steps will not serve to hinder or hold back any future upgrade.

Principle 3: Reinforcing the resilience of the network

118. The third principle that lies behind our vision, as set out in paragraph 63, is as follows: policy in this area should strive to reinforce the robustness and resilience of the network as a whole.

119. We note with interest that in the mid Nineteenth Century, a Select Committee set up to consider the UK's contemporary transport networks subscribed to a similar principle. Chaired by Viscount Cardwell, their report recommended that:

"Parliament should secure freedom and economy of transit from one end of the kingdom to the other and should compel railway companies to give the public the full advantage of convenient interchange from one railway system to another".

120. This Committee's work, the essence of which is encapsulated in this statement, paved the way for the Railway and Canal Act 1854.

121. At present, data in the UK seldom takes the logically most efficient course from A to B. Traffic from Edinburgh to Edinburgh, for example, travels via Manchester, Skipton or London. This is because different proprietary networks erect barriers to data originating on one, travelling over the other, even if permitting traffic across networks would allow data to take a more rational route. We rather suspect that the Cardwell Committee would not have approved.

122. Aside from forcing data around irrational routes, the effect of these barriers to cooperation is also to render the whole network more vulnerable. Curiously, the architecture of the Internet provides robustness through redundancy. Whereas the tree architecture of the legacy telephone network makes whole communities vulnerable to single points of failure,[74] a network with built-in redundancy which allows data flexibly to travel over any of part of it, would increase the robustness and resilience of the whole. As people become more reliant on broadband for day to day services, this will become ever more important.

123. Again, if the work of policy development in this area were to begin by asking what the UK's broadband infrastructure should look like, we would argue that the robustness and resilience of the network would emerge as a fundamental principle—and, therefore, ought to be an explicit component of policy.

124. It should be a fundamental principle of broadband policy that measures be undertaken, where possible, to reinforce the robustness and resilience of the network as a whole.

125. We recommend the Government ensure freedom and economy of passage for communication of data across the UK.

126. We recommend that Ofcom, in addition to its duties on competition and investment, be given an additional duty to monitor and foster the efficient utilisation of existing capacity (including, for example, use of the communications infrastructure owned by other infrastructure providers) to provide a robust and resilient national network that promotes affordable open access to wholesale and retail connectivity across the UK.

127. In outlining these three principles—driving broadband policy to reduce the digital divide, taking a long-term, but flexible view, and reinforcing the resilience of the network—we have clarified the significance of the differences between our vision, outlined at the start of this chapter, and the Government's. As has been made clear, many of our differences could have been avoided if the Government had started its policy thinking in the right place, namely by asking: what should the UK's broadband infrastructure look like in view of its place as a major strategic asset, at least on a par with the UK's roads, railways and energy networks?

Additional points relating to the Government's vision

128. In addition to clarifying and proposing a reorientation of the very vision behind Government policy, there are three other issues which the Government could usefully take into account in light of its current policy:

· Continuous upgrade and average speeds;

· Universal service obligation;

· Uptake and effective use of the internet;

129. First, however much we would resist them, the focus placed by the current policy on specific speed targets raises the importance of measurement. Currently, the scorecard which Ofcom proposes to use to determine the Government's success in achieving their targets appears to include measures of average speed. This is a problem.

130. Evaluating speeds which end-users receive on the basis of an average across a population is fraught with problems:

· It is difficult to know whether any increase in the average results from improvements for the 50% with speeds greater than the median;

· By extension, it is impossible to know whether higher average speeds are an indication of a narrower digital divide or a wider one.

131. Furthermore, because the distribution of speeds is highly skewed, the median is a more informative statistic.[75]

132. We recommend that the Government's targets should refer to minimum and median levels of service, and that Ofcom adapts its scorecard accordingly.

133. By extension, in order to ensure that the digital divide is braced and gradually drawn in, it will be important for the UK's low-end speeds to increase in line with the increase in median speeds. Otherwise, those with the slowest connections will not share in the advances elsewhere, and we will create a widening digital divide which will be socially divisive.

134. In order to ensure the digital divide is not widened, we recommend that the Government commit to reducing the digital divide between the minimal service levels guaranteed to all and the median service levels enjoyed by the majority.

135. Second, an unavoidable point of debate in connection with the stipulation of specific speeds and minimum service levels is whether it might be appropriate to introduce a Universal Service Obligation (USO) to oblige infrastructure providers and ISPs to provide both specific speeds and minimum service levels.

136. It is our view that a Universal Service Obligation (USO) is not an appropriate way to bring about universal access to minimum levels of service, not least because in practice, imposing legal obligations on ISPs could easily and quickly lead to drawn out proceedings in the courts.

137. We do, therefore, endorse the approach adopted by the Government: pledging a Universal Service Commitment, to which it will be politically accountable, and stating explicitly a clear political aspiration to provide universal access to a minimum level of broadband provision. This, in our view, is at this stage a more appropriate approach than introducing a legally-binding USO.

138. However, there is an important, and foreseeable, set of circumstances in which we would be likely to express an alternative view on this point. A view which appears to be commonly shared by the major ISPs is that, initially, the strongest driver of consumers moving from basic to enhanced broadband will be Internet Protocol Television (IPTV) services. The widespread involvement in YouView and the recent battle between BSkyB and BT Vision for rights to broadcast Premier League football games has only reinforced this point.

139. It is likely that IPTV services will become ever more widespread, and eventually the case for transferring the carriage of broadcast content, including public service broadcasting, from spectrum to the internet altogether will become overwhelming. This may well be a more sensible arrangement, as spectrum is perfectly suited to mobile applications, as Richard Hooper, OBE, Chairman of the Broadband Stakeholder Group, told us:

"Most people watch their television in fixed locations from fixed sets. Actually, spectrum's great wonder is its ability for mobility."[76]

140. As such, it might be argued that spectrum's current use for fixed, broadcast purposes is wasteful.

141. We recommend that the Government, Ofcom and the industry begin to consider the desirability of the transfer of terrestrial broadcast content from spectrum to the internet and the consequent switching off of broadcast transmission over spectrum, and in particular what the consequences of this might be and how we ought to begin to prepare.

142. As and when this occurs, and particularly if Public Service Broadcasting channels begin to be delivered primarily through the internet, the case for a USO, echoing that for television and radio, will become, in our view, significantly stronger.

143. While we do not support the introduction of a USO at present, we do believe that broadcast media will increasingly come to be delivered via the internet. As and when that happens, and particularly in circumstances where this applies to PSB channels, the argument for recommending a USO becomes stronger. The Government should begin now to give this active consideration.

144. Third, 8.12 million or 16.1% of adults in the UK have never been online[77] and many of these people are from some of the most vulnerable social groups, as Dido Harding, CEO of TalkTalk and board member of GO ON UK, underlined in oral evidence: "older, disabled or people for whom English is not their first language—those are the three big groups" and "the most digitally excluded people in society."[78]

145. These people are clearly at a manifest and material disadvantage. According to a report prepared by PwC for UK Digital Champion, Martha Lane Fox, there are a number of direct benefits to individuals of going online, including:

· Saving £560 a year by shopping and paying bills online;

· Better educational opportunities: "access to a computer and the internet can improve children's educational performance";

· Better employment opportunities: "people with good ICT skills earn between 3% and 10% more than people without such skills."[79]

146. Equally, moving people online is very much in the wider interest of the UK economy and the Government. According to the same report: "each contact and transaction with Government switched online could generate savings of between £3.30 and £12.00"[80] Evidence received from Arqiva, a media infrastructure and technology company, cast a spotlight on "studies undertaken by McKinsey, Allen, OECD and the World Bank [showing] that a 10% increase in broadband penetration results in a 1% increase in the rate of growth of GDP."[81]

147. In this inquiry we have focused on broadband infrastructure rather than digital literacy, although we may well look at this area more closely in the future. We are aware, however, that the achievements of Race Online 2012, the predecessor of GO ON UK, include helping to drive the number of people who have never been online down from 10 million to the current figure of 8 million over 2 years from 2010 to 2012. With further support, it may be possible to drive this down still further. In particular, we have heard a number of calls from those working in this area for the Government to elevate digital literacy and internet uptake to a higher level of priority. For example, Dido Harding told us:

"Government needs to explicitly set out that getting everyone online is a policy priority and therefore drive further activity in this area."[82]

148. We have also heard of an innovative scheme run by Three promoting digital inclusion:

"We have been looking at identifying organisations and individuals who have contact with people likely to be socially, economically and digitally excluded. These people are already delivering services. We tried to encourage them to use our connectivity in their service provision. We give that connectivity for free to those organisations and individuals so that they can incorporate it into the work that they do, so it does not become, 'Come along and have a six-week taster course to enjoy the benefits of the internet'. At times, that can be quite off-putting. It just becomes part of the service that you are already accessing—be it the library or hospice-at-home service, where the volunteers sit with the individual for a number of hours, keeping them company. It is about making it relevant to that individual."[83]

149. We welcome this initiative and hope that similar initiatives can be devised and extended in the future.

We bring to the Government's attention the fact that we have heard a number of calls during this inquiry, with which we agree, for uptake and effective use of the internet to become a higher priority.

42 DCMS and BIS, Britain's Superfast Broadband Future, December 2010. Available online: http://www.culture.gov.uk/images/publications/10-1320-britains-superfast-broadband-future.pdf Back

43 ibid. Back

44 ibid. Back

45 Though currently unlikely, it may be a matter of future welfare policy to ensure that everyone can afford to do so. Also see paragraphs 65 and 150. Back

46 DCMS press release, 'Next phase of superfast broadband plans announced,' 6 December 2010. Available online: http://www.culture.gov.uk/news/media_releases/7619.aspx Back

47 Q 767 Back

48 Q 807 Back

49 Guardian, 'UK rural broadband on hold as European commissioners dig in heels', 3 July 2012. Available online: http://www.guardian.co.uk/technology/2012/jul/03/bduk-broadband-scheme-delayed-european-commission Back

50 EU Commission press release, 'State aid: Commission consults on draft guidelines for broadband networks', 1 June 2012. Available online: http://europa.eu/rapid/pressReleasesAction.do?reference=IP/12/550&format=HTML&aged=0&language=EN&guiLanguage=en Back

51 Paras 67(g) and 76(b). The consultation document is available from the link in footnote 50 above. Back

52 European Commission-policy statement by Vice-President Kroes, 'Enhancing the broadband investment environment', 12 July 2012. Available online: http://europa.eu/rapid/pressReleasesAction.do?reference=MEMO/12/554&format=HTML&aged=0&language=EN&guiLanguage=en Back

53 DCMS and BIS, Britain's Superfast Broadband Future, December 2010. Available online: http://www.culture.gov.uk/images/publications/10-1320-britains-superfast-broadband-future.pdf Back